Stretching the life of Twitter classifiers with time-stamped semantic graphs

Social media has become an effective channel for communicating both trends and public opinion on current events. However the automatic topic classification of social media content pose various challenges. Topic classification is a common technique used for automatically capturing themes that emerge from social media streams. However, such techniques are sensitive to the evolution of topics when new event-dependent vocabularies start to emerge (e.g., Crimea becoming relevant to War Conflict during the Ukraine crisis in 2014). Therefore, traditional supervised classification methods which rely on labelled data could rapidly become outdated. In this paper we propose a novel transfer learning approach to address the classification task of new data when the only available labelled data belong to a previous epoch. This approach relies on the incorporation of knowledge from DBpedia graphs. Our findings show promising results in understanding how features age, and how semantic features can support the evolution of topic classifiers.

Every hierarchy of beliefs is a type

When modeling game situations of incomplete information one usually considers the players’ hierarchies of beliefs, a source of all sorts of complications. Harsányi (1967-68)’s idea henceforth referred to as the ”Harsányi program” is that hierarchies of beliefs can be replaced by ”types”. The types constitute the ”type space”. In the purely measurable framework Heifetz and Samet (1998) formalize the concept of type spaces and prove the existence and the uniqueness of a universal type space. Meier (2001) shows that the purely measurable universal type space is complete, i.e., it is a consistent object. With the aim of adding the finishing touch to these results, we will prove in this paper that in the purely measurable framework every hierarchy of beliefs can be represented by a unique element of the complete universal type space.

Statistical Analysis of Text Summarization Evaluation

This dissertation applies statistical methods to the evaluation of automatic summarization using data from the Text Analysis Conferences in 2008-2011. Several aspects of the evaluation framework itself are studied, including the statistical testing used to determine significant differences, the assessors, and the design of the experiment. In addition, a family of evaluation metrics is developed to predict the score an automatically generated summary would receive from a human judge and its results are demonstrated at the Text Analysis Conference. Finally, variations on the evaluation framework are studied and their relative merits considered. An over-arching theme of this dissertation is the application of standard statistical methods to data that does not conform to the usual testing assumptions.

A system’s approach to cache hierarchy-aware decomposition of data-parallel computations

Dissertação para obtenção do Grau de Mestre em Engenharia Informática

Construction of an index of information from clinical practice in Radiology and Imaging Diagnosis based on text mining and thesaurus

Objective To construct a Portuguese language index of information on the practice of diagnostic radiology in order to improve the standardization of the medical language and terminology. Materials and Methods A total of 61,461 definitive reports were collected from the database of the Radiology Information System at Hospital das Clínicas – Faculdade de Medicina de Ribeirão Preto (RIS/HCFMRP) as follows: 30,000 chest x-ray reports; 27,000 mammography reports; and 4,461 thyroid ultrasonography reports. The text mining technique was applied for the selection of terms, and the ANSI/NISO Z39.19-2005 standard was utilized to construct the index based on a thesaurus structure. The system was created in *html. Results The text mining resulted in a set of 358,236 (n = 100%) words. Out of this total, 76,347 (n = 21%) terms were selected to form the index. Such terms refer to anatomical pathology description, imaging techniques, equipment, type of study and some other composite terms. The index system was developed with 78,538 *html web pages. Conclusion The utilization of text mining on a radiological reports database has allowed the construction of a lexical system in Portuguese language consistent with the clinical practice in Radiology.

Credibility, rules and power in the European Union institutions: a transactional analysis of the "Stability and Growth pact"

The credibility of the rules and the elements of power constitute fundamental keys in the analysis of the political institutions. This paper opens the "black box" of the European Union institutions and analyses the problem of credibility in the commitment of the Stability and Growth pact (SGP). This Pact (SGP) constituted a formal rule that tried to enforce budgetary discipline on the European States. Compliance with this contract could be ensured by the existence of "third party enforcement" or by the coincidence of the ex-ante and ex-post interests of the States (reputational capital). The fact is that states such as France or Germany failed to comply with the ruling and managed to avoid the application of sanctions. This article studies the transactions and the hierarchy of power that exists in the European institutions, and analyses the institutional framework included in the new European Constitution.

The Impact of a Text-Reader Support Program on Intermediate Students' Reading Comprehension

This study examined the efficacy of providing four Grade 7 and 8 students with reading difficulties with explicit instruction in the use of reading comprehension strategies while using text-reader software. Specifically, the study explored participants' combined use of a text-reader and question-answering comprehension strategy during a 6-week instructional program. Using a qualitative case study methodology approach, participants' experiences using text-reader software, with the presence of explicit instruction in evidence-based reading comprehension strategies, were examined. The study involved three phases: (a) the first phase consisted of individual interviews with the participants and their parents; (b) the second phase consisted of a nine session course; and (c) the third phase consisted of individual exit interviews and a focus group discussion. After the data collection phases were completed, data were analyzed and coded for emerging themes, with-quantitativ,e measures of participants' reading performance used as descriptive data. The data suggested that assistive technology can serve as an instructional "hook", motivating students to engage actively in the reading processes, especially when accompanied by explicit strategy instruction. Participants' experiences also reflected development of strategy use and use of text-reader software and the importance of social interactions in developing reading comprehension skills. The findings of this study support the view that the integration of instruction using evidence-based practices are important and vital components in the inclusion oftext-reader software as part of students' educational programming. Also, the findings from this study can be extended to develop in-class programming for students using text-reader software.

Model Selection in Summary Evaluation

A difficulty in the design of automated text summarization algorithms is in the objective evaluation. Viewing summarization as a tradeoff between length and information content, we introduce a technique based on a hierarchy of classifiers to rank, through model selection, different summarization methods. This summary evaluation technique allows for broader comparison of summarization methods than the traditional techniques of summary evaluation. We present an empirical study of two simple, albeit widely used, summarization methods that shows the different usages of this automated task-based evaluation system and confirms the results obtained with human-based evaluation methods over smaller corpora.

Dynamic management and restoration of virtual paths in broadband networks based on distributed software agents

La gestió de xarxes és un camp molt ampli i inclou molts aspectes diferents. Aquesta tesi doctoral està centrada en la gestió dels recursos en les xarxes de banda ampla que disposin de mecanismes per fer reserves de recursos, com per exemple Asynchronous Transfer Mode (ATM) o Multi-Protocol Label Switching (MPLS). Es poden establir xarxes lògiques utilitzant els Virtual Paths (VP) d'ATM o els Label Switched Paths (LSP) de MPLS, als que anomenem genèricament camins lògics. Els usuaris de la xarxa utilitzen doncs aquests camins lògics, que poden tenir recursos assignats, per establir les seves comunicacions. A més, els camins lògics són molt flexibles i les seves característiques es poden canviar dinàmicament. Aquest treball, se centra, en particular, en la gestió dinàmica d'aquesta xarxa lògica per tal de maximitzar-ne el rendiment i adaptar-la a les connexions ofertes. En aquest escenari, hi ha diversos mecanismes que poden afectar i modificar les característiques dels camins lògics (ample de banda, ruta, etc.). Aquests mecanismes inclouen els de balanceig de la càrrega (reassignació d'ample de banda i reencaminament) i els de restauració de fallades (ús de camins lògics de backup). Aquests dos mecanismes poden modificar la xarxa lògica i gestionar els recursos (ample de banda) dels enllaços físics. Per tant, existeix la necessitat de coordinar aquests mecanismes per evitar possibles interferències. La gestió de recursos convencional que fa ús de la xarxa lògica, recalcula periòdicament (per exemple cada hora o cada dia) tota la xarxa lògica d'una forma centralitzada. Això introdueix el problema que els reajustaments de la xarxa lògica no es realitzen en el moment en què realment hi ha problemes. D'altra banda també introdueix la necessitat de mantenir una visió centralitzada de tota la xarxa. En aquesta tesi, es proposa una arquitectura distribuïda basada en un sistema multi agent. L'objectiu principal d'aquesta arquitectura és realitzar de forma conjunta i coordinada la gestió de recursos a nivell de xarxa lògica, integrant els mecanismes de reajustament d'ample de banda amb els mecanismes de restauració preplanejada, inclosa la gestió de l'ample de banda reservada per a la restauració. Es proposa que aquesta gestió es porti a terme d'una forma contínua, no periòdica, actuant quan es detecta el problema (quan un camí lògic està congestionat, o sigui, quan està rebutjant peticions de connexió dels usuaris perquè està saturat) i d'una forma completament distribuïda, o sigui, sense mantenir una visió global de la xarxa. Així doncs, l'arquitectura proposada realitza petits rearranjaments a la xarxa lògica adaptant-la d'una forma contínua a la demanda dels usuaris. L'arquitectura proposada també té en consideració altres objectius com l'escalabilitat, la modularitat, la robustesa, la flexibilitat i la simplicitat. El sistema multi agent proposat està estructurat en dues capes d'agents: els agents de monitorització (M) i els de rendiment (P). Aquests agents estan situats en els diferents nodes de la xarxa: hi ha un agent P i diversos agents M a cada node; aquests últims subordinats als P. Per tant l'arquitectura proposada es pot veure com una jerarquia d'agents. Cada agent és responsable de monitoritzar i controlar els recursos als que està assignat. S'han realitzat diferents experiments utilitzant un simulador distribuït a nivell de connexió proposat per nosaltres mateixos. Els resultats mostren que l'arquitectura proposada és capaç de realitzar les tasques assignades de detecció de la congestió, reassignació dinàmica d'ample de banda i reencaminament d'una forma coordinada amb els mecanismes de restauració preplanejada i gestió de l'ample de banda reservat per la restauració. L'arquitectura distribuïda ofereix una escalabilitat i robustesa acceptables gràcies a la seva flexibilitat i modularitat.

The basic ingredients of the North Atlantic storm track. Part I: land-sea contrast and orography

Understanding and predicting changes in storm tracks over longer time scales is a challenging problem, particularly in the North Atlantic. This is due in part to the complex range of forcings (land–sea contrast, orography, sea surface temperatures, etc.) that combine to produce the structure of the storm track. The impact of land–sea contrast and midlatitude orography on the North Atlantic storm track is investigated through a hierarchy of GCM simulations using idealized and “semirealistic” boundary conditions in a high-resolution version of the Hadley Centre atmosphere model (HadAM3). This framework captures the large-scale essence of features such as the North and South American continents, Eurasia, and the Rocky Mountains, enabling the results to be applied more directly to realistic modeling situations than was possible with previous idealized studies. The physical processes by which the forcing mechanisms impact the large-scale flow and the midlatitude storm tracks are discussed. The characteristics of the North American continent are found to be very important in generating the structure of the North Atlantic storm track. In particular, the southwest–northeast tilt in the upper tropospheric jet produced by southward deflection of the westerly flow incident on the Rocky Mountains leads to enhanced storm development along an axis close to that of the continent’s eastern coastline. The approximately triangular shape of North America also enables a cold pool of air to develop in the northeast, intensifying the surface temperature contrast across the eastern coastline, consistent with further enhancements of baroclinicity and storm growth along the same axis.

Identifying problematic classes in text classification

Real-world text classification tasks often suffer from poor class structure with many overlapping classes and blurred boundaries. Training data pooled from multiple sources tend to be inconsistent and contain erroneous labelling, leading to poor performance of standard text classifiers. The classification of health service products to specialized procurement classes is used to examine and quantify the extent of these problems. A novel method is presented to analyze the labelled data by selectively merging classes where there is not enough information for the classifier to distinguish them. Initial results show the method can identify the most problematic classes, which can be used either as a focus to improve the training data or to merge classes to increase confidence in the predicted results of the classifier.

Crystal design approaches for the synthesis of paracetamol co-crystals

Crystal engineering principles were used to design three new co-crystals of paracetamol. A variety of potential cocrystal formers were initially identified from a search of the Cambridge Structural Database for molecules with complementary hydrogen-bond forming functionalities. Subsequent screening by powder X-ray diffraction of the products of the reaction of this library of molecules with paracetamol led to the discovery of new binary crystalline phases of paracetamol with trans-1,4- diaminocyclohexane (1); trans-1,4-di(4-pyridyl)ethylene (2); and 1,2-bis(4-pyridyl)ethane (3). The co-crystals were characterized by IR spectroscopy, differential scanning calorimetry, and 1H NMR spectroscopy. Single crystal X-ray structure analysis reveals that in all three co-crystals the co-crystal formers (CCF) are hydrogen bonded to the paracetamol molecules through O−H···N interactions. In co-crystals (1) and (2) the CCFs are interleaved between the chains of paracetamol molecules, while in co-crystal (3) there is an additional N−H···N hydrogen bond between the two components. A hierarchy of hydrogen bond formation is observed in which the best donor in the system, the phenolic O−H group of paracetamol, is preferentially hydrogen bonded to the best acceptor, the basic nitrogen atom of the co-crystal former. The geometric aspects of the hydrogen bonds in co-crystals 1−3 are discussed in terms of their electrostatic and charge-transfer components.

The reversibility of sea level rise

During the last century, global climate has been warming, and projections indicate that such a warming is likely to continue over coming decades. Most of the extra heat is stored in the ocean, resulting in thermal expansion of seawater and global mean sea level rise. Previous studies have shown that after CO2 emissions cease or CO2 concentration is stabilized, global mean surface air temperature stabilizes or decreases slowly, but sea level continues to rise. Using idealized CO2 scenario simulations with a hierarchy of models including an AOGCM and a step-response model, the authors show how the evolution of thermal expansion can be interpreted in terms of the climate energy balance and the vertical profile of ocean warming. Whereas surface temperature depends on cumulative CO2 emissions, sea level rise due to thermal expansion depends on the time profile of emissions. Sea level rise is smaller for later emissions, implying that targets to limit sea level rise would need to refer to the rate of emissions, not only to the time integral. Thermal expansion is in principle reversible, but to halt or reverse it quickly requires the radiative forcing to be reduced substantially, which is possible on centennial time scales only by geoengineering. If it could be done, the results indicate that heat would leave the ocean more readily than it entered, but even if thermal expansion were returned to zero, the geographical pattern of sea level would be altered. Therefore, despite any aggressive CO2 mitigation, regional sea level change is inevitable.

Impacts of increasing the aerosol complexity in the Met Office global numerical weather prediction model.

The inclusion of the direct and indirect radiative effects of aerosols in high-resolution global numerical weather prediction (NWP) models is being increasingly recognised as important for the improved accuracy of short-range weather forecasts. In this study the impacts of increasing the aerosol complexity in the global NWP configuration of the Met Office Unified Model (MetUM) are investigated. A hierarchy of aerosol representations are evaluated including three-dimensional monthly mean speciated aerosol climatologies, fully prognostic aerosols modelled using the CLASSIC aerosol scheme and finally, initialised aerosols using assimilated aerosol fields from the GEMS project. The prognostic aerosol schemes are better able to predict the temporal and spatial variation of atmospheric aerosol optical depth, which is particularly important in cases of large sporadic aerosol events such as large dust storms or forest fires. Including the direct effect of aerosols improves model biases in outgoing long-wave radiation over West Africa due to a better representation of dust. However, uncertainties in dust optical properties propagate to its direct effect and the subsequent model response. Inclusion of the indirect aerosol effects improves surface radiation biases at the North Slope of Alaska ARM site due to lower cloud amounts in high-latitude clean-air regions. This leads to improved temperature and height forecasts in this region. Impacts on the global mean model precipitation and large-scale circulation fields were found to be generally small in the short-range forecasts. However, the indirect aerosol effect leads to a strengthening of the low-level monsoon flow over the Arabian Sea and Bay of Bengal and an increase in precipitation over Southeast Asia. Regional impacts on the African Easterly Jet (AEJ) are also presented with the large dust loading in the aerosol climatology enhancing of the heat low over West Africa and weakening the AEJ. This study highlights the importance of including a more realistic treatment of aerosol–cloud interactions in global NWP models and the potential for improved global environmental prediction systems through the incorporation of more complex aerosol schemes.